Extensive Cargo Identification Reveals Distinct Biological Roles of the 12
Total Page:16
File Type:pdf, Size:1020Kb
TOOLS AND RESOURCES Extensive cargo identification reveals distinct biological roles of the 12 importin pathways Makoto Kimura1*, Yuriko Morinaka1, Kenichiro Imai2,3, Shingo Kose1, Paul Horton2,3, Naoko Imamoto1* 1Cellular Dynamics Laboratory, RIKEN, Wako, Japan; 2Artificial Intelligence Research Center, National Institute of Advanced Industrial Science and Technology, Tokyo, Japan; 3Biotechnology Research Institute for Drug Discovery, National Institute of Advanced Industrial Science and Technology, Tokyo, Japan Abstract Vast numbers of proteins are transported into and out of the nuclei by approximately 20 species of importin-b family nucleocytoplasmic transport receptors. However, the significance of the multiple parallel transport pathways that the receptors constitute is poorly understood because only limited numbers of cargo proteins have been reported. Here, we identified cargo proteins specific to the 12 species of human import receptors with a high-throughput method that employs stable isotope labeling with amino acids in cell culture, an in vitro reconstituted transport system, and quantitative mass spectrometry. The identified cargoes illuminated the manner of cargo allocation to the receptors. The redundancies of the receptors vary widely depending on the cargo protein. Cargoes of the same receptor are functionally related to one another, and the predominant protein groups in the cargo cohorts differ among the receptors. Thus, the receptors are linked to distinct biological processes by the nature of their cargoes. DOI: 10.7554/eLife.21184.001 *For correspondence: makimura@ riken.jp (MK); [email protected] Introduction (NI) In interphase cells, proteins and RNAs migrate into and out of the nuclei through the central chan- Competing interests: The nels of the nuclear pore complexes (NPC) embedded in the nuclear envelope. These nuclear pores authors declare that no are lined with FG-repeat domains that constitute a permeability barrier, and only macromolecules competing interests exist. that reversibly interact with the FG-repeats can permeate this barrier (Schmidt and Go¨rlich, 2016). One such group of proteins, the importin (Imp)-b family proteins, are nucleocytoplasmic transport Funding: See page 23 receptors (NTRs) that primarily carry nuclear proteins and small RNAs as their cargoes through the Received: 02 September 2016 nuclear pores, although non-importin family NTRs also act depending on the cargo and physiological Accepted: 17 January 2017 conditions (Kose et al., 2012; Lu et al., 2014; Weberruss et al., 2013). The human genome enco- Published: 24 January 2017 des 20 species of Imp-b family NTRs, of which 10 [Imp-b, transportin (Trn)-1, -2, -SR (-3), Imp-4, -5 Reviewing editor: Karsten Weis, (RanBP5), -7, -8, -9, and -11] are nuclear import receptors, 7 [exportin (Exp)-1 (CRM1), -2 (CAS/ ETH Zurich, Switzerland CSE1L), -5, -6, -7, -t, and RanBP17] are export receptors, 2 (Imp-13 and Exp-4) are bi-directional receptors, and the function of RanBP6 is undetermined (Kimura and Imamoto, 2014). These NTRs Copyright Kimura et al. This constitute multiple parallel transport pathways. The basic mechanism of directional transport was article is distributed under the elucidated in the early years (Go¨rlich and Kutay, 1999), but even today, the number of NTR-specific terms of the Creative Commons Attribution License, which cargoes that has been reported is surprisingly small, hindering a biological understanding of the permits unrestricted use and nucleocytoplasmic transport system. redistribution provided that the NTRs are thought to transport specific cohorts of cargoes by binding to specific sites on those original author and source are cargoes (Chook and Su¨el, 2011), but the consensus structures of the NTR-binding sites have been credited. established for only a few NTRs (Soniat and Chook, 2015) as follows: the classical nuclear Kimura et al. eLife 2017;6:e21184. DOI: 10.7554/eLife.21184 1 of 31 Tools and resources Biochemistry Cell Biology localization signal (cNLS) for the Imp-a family adapters, which connects Imp-b and cargoes (Lange et al., 2007); PY-NLS for Trn-1 and -2 (Lee et al., 2006; Su¨el et al., 2008); the nuclear export signal (NES) for Exp-1 (Hutten and Kehlenbach, 2007); the SR-rich domain that binds to Trn-SR (Kataoka et al., 1999; Maertens et al., 2014); and Lys-rich NLS (IK-NLS) for yeast Kap121p (Imp-5 homolog; Kobayashi and Matsuura, 2013; Kobayashi et al., 2015). The b-like importin-binding (BIB) domain is another NTR-binding site (Ja¨kel and Go¨rlich, 1998), but its consensus sequence and NTR specificity remain obscure. Among the NTRs, Imp-b exclusively uses one of the seven species of the Imp-a family of proteins as an adapter for cargo binding, and many Imp-a/b cargoes have been reported, although Imp-b also directly binds to cargoes (Goldfarb et al., 2004). Among the import receptors, Trn-1 and its closest homolog Trn-2 have the second-highest number of cargoes reported thus far, and the PY-NLS motif has been defined, although in some cases the motif is difficult to rec- ognize because of sequence diversity and structural disorder is another requisite (Soniat and Chook, 2015, 2016). For the cargoes of other NTRs, the consensus structures of NTR-binding sites have hardly been derived because only limited numbers of cargoes have been reported, including Imp-b- direct cargoes. There are many reports on the differential spatiotemporal expression of Imp-b family NTRs, including tissue specificities in humans (Quan et al., 2008), developmental or spermatogenic stage specificities in mice (Major et al., 2011; Quan et al., 2008), and tissue or response specificities in plants (Huang et al., 2010). Expression regulation is not only transcriptional but also miRNA-medi- ated (Li et al., 2013; Szczyrba et al., 2013) or locally translationally mediated (Perry and Fainzilber, 2009). Additionally, the NTRs are functionally regulated by protein modifications (Wang et al., 2009), inhibitory factors (Lieu et al., 2014), and specific anchorings (Makhnevych et al., 2003). These nucleocytoplasmic transport regulations must significantly influence cellular physiology, and their significance may be elucidated if the affected cargoes can be specified. Indeed, in previous studies, NTR regulations have been linked to cellular responses through the functions of specific car- goes. For example, in prostate cancer cells treated with a cinnamaldehyde derivative, the expression of Imp-7 and the transcription factor Egr1 are induced, and the Egr1 imported by Imp-7 activates apoptotic gene transcription (Kang et al., 2013). In another example, when the nuclear import of some ribosomal proteins (RPs) is inhibited by the repression of Imp-7 expression, other unassembled RPs restrain the negative regulator of p53 Mdm2 and thereby activate p53 to inhibit cell growth (Golomb et al., 2012). Additionally, the inhibition of Trn-2 by the caspase-generated HuR (ELAVL1) fragment is crucial for the cytoplasmic retention of full-length HuR, which induces myogenesis (Beauchamp et al., 2010) or staurosporine-induced apoptosis (von Roretz et al., 2011). In many other studies, mutations of particular NTR genes in model organisms, including yeast, flies, and plants, have resulted in defects in specific biological processes (Kimura and Imamoto, 2014). Thus, each NTR has its own inherent biological significance. However, the details of the molecular pro- cesses are largely uncharacterized because the responsible cargoes have not been identified. If we could identify more cargoes, further studies of cellular regulation by nucleocytoplasmic transport would be possible. We previously established a method for identifying the cargoes of a nuclear import receptor called SILAC-Tp (Kimura et al., 2013a, 2013b, 2014). SILAC-Tp employs stable isotope labeling with amino acids in cell culture (SILAC) (Ong et al., 2002), an in vitro reconstituted nuclear transport system (Adam et al., 1990), and quantitative mass spectrometry. A recent advancement of the Orbi- trap mass spectrometer drastically increased the identified and quantified protein numbers, and this advancement has been successfully applied to other cargo identification methods (Kırlı et al., 2015; Thakar et al., 2013). Here, we utilized this advancement for the SILAC-Tp method and identified import cargoes of all 12 NTRs, of which 10 are import and two are bi-directional receptors. Our results illustrate the basic framework and the biological significance of the nucleocytoplasmic trans- port pathways. Results and discussion SILAC-Tp effectively identifies cargoes SILAC-Tp employs an in vitro nuclear transport system, and all 12 NTRs import their reported spe- cific cargoes in this system (Figure 1—figure supplement 1C). The transport system consists of Kimura et al. eLife 2017;6:e21184. DOI: 10.7554/eLife.21184 2 of 31 Tools and resources Biochemistry Cell Biology permeabilized HeLa cells labeled with ‘heavy’ amino acids by SILAC, unlabeled HeLa nuclear extract depleted of Imp-b family NTRs and RCC1, unlabeled HeLa cytosolic extract depleted of Imp-b family NTRs, one species of recombinant NTR, p10/NTF2, and an ATP regeneration system. Unlabeled ‘light’ proteins in the nuclear extract are imported into the nuclei of the permeabilized cells. Simulta- neously, a control reaction without the NTR is performed. Next, the proteins are extracted from the nuclei and identified and quantified by LC-MS/MS. The recipient nuclei contain both the imported and endogenous proteins, and the ratio of the imported to the endogenous fraction of a protein is calculated as the unlabeled/labeled or light/heavy (L/H) ratio. The quotient of the L/H ratios with the NTR (+NTR) and without it (control), that is, (L/H+NTR)/(L/HCtl), of a protein is defined as the +NTR/ Ctl value and is used as the index for cargo potentiality. In one run of SILAC-Tp (control or +NTR), approximately 2500 to 4000 proteins were identified, and the L/H ratios of 1700 to 3100 proteins were quantified. To calculate the +NTR/Ctl value, one protein has to be quantified in both the control and +NTR reactions, and we discarded L/H+NTR val- ues that lacked the counterpart L/HCtl values.